The long-term goal of this proposal is to define the molecular parameters that mediate oncogenic transformation of acute myeloid leukemia stem cells (LSCs). This objective is based on the overwhelming evidence that now exists regarding the role and biological properties of LSCs. Numerous studies have documented the existence of LSCs in both primary human tissues and mouse model systems. This biologically distinct subpopulation is responsible for the genesis and perpetuation of leukemic disease. Further, LSCs display significance resistance to conventional forms of therapy, making the development of novel regimens a high priority. For this application we have employed a novel strategy for the analysis of oncogene cooperativity as a means to elucidate the genes/pathways most central to malignancy. The approach employs global gene expression profiling as a tool to identify genes that are synergistically dysregulated as a consequence of co- expressing two cooperating oncogenes. Previous studies in a model of epithelial cancer have demonstrated that this strategy is highly effective, and indeed successfully yielded a number of novel mechanisms involved in tumor formation. We have applied this approach, known as identification of cooperativity response genes or CRGs in the context of a genetically defined murine model of blast crisis CML. The system, which employs dual expression of the BCR/ABL and Nup98/HoxA9 translocation products, has previously been characterized in several reports, and has a phenotypically defined LSC population. Our experiments have identified 72 genes that fulfill the CRG criteria in this model. Preliminary studies of one CRG, serpinB2, have indicated an important role for the gene in growth of leukemic cells in vivo. Further, analysis of the CRG expression profile has been performed using computational methods, which indicate that ERBB2 signaling may be important for leukemia cell growth/survival. Initial studies with a drug that targets ERBB2 have indicated a functional role for this pathway. Taken together, our preliminary indicate that CRG-based analysis of leukemia may be a powerful means by which to identify the mechanisms central to leukemogenesis. Based on the findings to date, we hypothesize that CRGs are highly enriched for critical regulators of leukemogenesis in our mouse model system. Further, we propose to investigate whether such genes are also important for the growth of primary human blast crisis CML. For both mouse and human studies, our emphasis will be on the analysis of CRGs in leukemic stem cell populations. Taken together, these studies will provide a comprehensive understanding of pathways and processes that mediate leukemic disease.